Extended Summaries ± Neurotox '98
a-BgTx, EPI or IMI. In contrast the dissociation rate
was markedly increased when initiated by MLA
compared to a-BgTx, EPI or IMI (Fig 1).
Displacement of [125I]a-BgTx, [3H]EPI and
3 Ingeborg van den Beukel,1* Remco Klaassen,2
[ H]IMI by MLA was with high potency with an
IC50 value close to the Kd value of each of these
labelled ligands. Displacement of [3H]MLA by a-
BgTx, EPI and IMI resulted in low potency, but was
characterised by shallow displacement curves.
4 DISCUSSION AND CONCLUSIONS
Saturation studies in M persicae membranes demon-
strate that [125I]a-BgTx, [3H]EPI, and [3H]IMI all
label two binding components of differing af®nities in
a ratio of 1:3. In each case the sum of the high and low
af®nity Bmax values approximated the Bmax value of the
single site labelled by [3H]MLA. [3H]MLA is a novel
radioligand which is highly selective in vertebrate
nAChRs,3 but which appears to be unable to
distinguish between a heterogeneous population of
nicotinic ligand binding sites in insects.
The increase in dissociation rate of [3H]a-BgTx
using MLA instead of either a-BgTx, EPI or IMI is
consistent with an allosteric interaction by MLA at a
binding site other than that of the high af®nity [3H]a-
BgTx binding site.
Displacement studies have shown that MLA is a
potent ligand at sites labelled with high af®nity by
[125I]a-BgTx, [3H]EPI, and [3H]IMI. The shallow
displacement curves obtained when [3H]MLA is
displaced by a-BgTx, EPI, and IMI (data not shown)
are likely to be due to displacement of [3H]MLA from
at least two distinct binding sites, for which the
displacing ligands have different af®nities, but which
are indistinguishable by [3H]MLA.
Interestingly, the high af®nity binding of [3H]IMI in
hemipteran membranes compared to that in non-
hemipteran insects may help explain why IMI is
particularly useful for the control of sucking pests.
ACKNOWLEDGEMENTS
Thanks to Matt Cahill of IACR Rothamsted for
supplying B tabaci. The work was supported by the
BBSRC and Zeneca Agrochemicals.
REFERENCES
1 Elbert A, Becker B, Hartwig J and Erdelen C, Imidacloprid ± a new
systemic insecticide. P¯anzenschutz-Nachrichten Bayer 44:113±
136 (1991).
2 Lind RJ, Clough MS, Reynolds SE and Earley FGP, [3H]Imida-
cloprid labels high and low af®nity nicotinic acetylcholine
receptor-like binding sites in the aphid Myzus persicae (Hemi-
ptera: Aphididae). Pestic Biochem Physiol 62:3±14 (1998).
3 Davies ARL, Hardick DJ, Blagbrough IS, Potter BVL, Wolsten-
holme AJ and Wonnacott S, [3H]Methyllycaconitine a novel
radioligand for a7-type nicotinic acetylcholine receptors. Soc
Neurosci Abstr 23:477.2 (1997).
Pestic Sci 55:1027±1040 (1999)
Nicotinic acetylcholine receptor chimeras of rat
a7 and Drosophila SAD reveal species-specific
agonist binding regions
Guus B Smit,2 Regina GDM van Kleef1 and
Marga Oortgiesen1†
1
Research Institute of Toxicology, Utrecht University, PO Box
80176, 3508 TD Utrecht, The Netherlands
2
Department of Molecular Neurobiology, Graduate School of
Neuroscience, Free University, De Boelelaan 1087, 1081 HV
Amsterdam, The Netherlands
Abstract: Species-speci®c agonist binding regions of
nicotinic acetylcholine receptors (nAChR) were
examined. Imidacloprid and physostigmine (Phy)
selectively activated insect nAChR composed of
Drosophila second alpha-like subunit (SAD) and
chick b2, in contrast to rat a7 nAChR. The Phy-
activated currents were a-bungarotoxin (a-BGT)
sensitive, suggesting activation at the agonist bind-
ing loop C. Several SAD-a7 chimeras were con-
structed, by switching agonist binding regions, and
expressed in oocytes. Though none of the chimeras
was activated by a range of nicotinic agonists,
[125I]a-BGT binding revealed homomeric assembly
of all chimeric cDNAs. Phy differentially displaced
[125I]a-BGT from the nAChR chimeras, suggesting
that the b subunit is not involved in Phy binding, and
that Phy targets the insect agonist binding loop C.
Keywords: nicotinic receptor; agonist binding;
chimera; physostigmine; imidacloprid; acetylcholine
receptor; neonicotinoid
1 INTRODUCTION
Multiple nicotinic acetylcholine receptor (nAChR)
subtypes exist, both within and between species.
Insects express only neuronal nAChR, while, in
vertebrates, endplate and neuronal nAChR types are
distinguished. A range of different insect and verte-
brate a subunits exist, which may be combined with
various b subunits to form functional nAChR. Three
loops in the a subunits are thought to be involved in
acetylcholine (ACh) binding,1 ie loop A (residues 86±
93), loop B (residues 148±151), and loop C (residues
190±198). The latter includes the two adjacent
cysteines characteristic of a subunits. Distinct nAChR
subtypes are distinguished by their physiological and
pharmacological properties. In addition, differential
sensitivities to chemical compounds may occur,
suggesting a role in species-selective toxicity. The
nitroguanidine insecticide imidacloprid selectively
activates insect nAChR by binding to the agonist
* Correspondence to: Ingeborg van den Beukel, Research Institute
of Toxicology, Utrecht University, PO Box 80176, 3508 Utrecht,
The Netherlands
E-mail: I.vandenBeukel@ritox.dgk.ruu.nl
†
Present address: Department of Neurobiology, Duke University
Medical Center, 435 Bryan Research Building, Durham, NC 27278,
USA
(Received 20 August 1998; revised version received 25 November
1998; accepted 20 May 1999)
1031
Extended Summaries ± Neurotox '98

binding site.2,3 The carbamate, physostigmine (Phy),

selectively activates whole-cell nicotinic currents in
locust neurons and in Xenopus oocytes expressing fruit
¯y nAChR resulting from cDNA injection of the
second a-like Drosophila (SAD) and the chick b2
subunit.4 Whole-cell currents during stimulation with
Phy could not be detected in various cell types
expressing different vertebrate nAChR.4 Phy appears
also to target the nAChR agonist binding site, as Phy-
activated currents are sensitive to competitive antago-
nists.4 This contrasts with previous reports that Phy
binds to a separate site instead of the agonist binding
site at vertebrate nAChR.5 These observations suggest
that ACh binding sites of insects differ from those of
vertebrates.
We investigated species-speci®c nAChR agonist
binding sites at the cellular and molecular level.
Several nAChR chimeras of rat a7 and Drosophila
SAD were constructed, in which distinct agonist
binding domains of a7 were replaced by the corre-
sponding domains of SAD cDNA. Agonist binding
and function of rat a7, insect SAD/chick b2, and the
chimeras in Xenopus oocytes was analysed.

Figure 1. Properties of oocytes expressing SAD/chick b2, a7 and a7-SAD

2 METHODS AND RESULTS
Pharmacological properties of rat a7 nAChR and fruit
¯y nAChR resulting from co-expression of the
Drosophila SAD a and the chick b2 subunit in Xenopus
oocytes were investigated, using the dual microelec-
trode voltage clamp method. ACh activated the
wildtype rat a7 and SAD/chick b2 nAChR. Con-
centration-effect curves revealed EC50 values of
179 (65) mM (n = 3) and 15 (4) mM (n = 3), with Hill
coef®cients of 1.44 (0.22) and 1.35 (0.09), respec-
tively, which are in the same range as those reported
previously.6,7 Imidacloprid at 10 mM (data not shown),
and 100 mM Phy evoked inward currents in oocytes
expressing SAD/chick b2 nAChR (Fig 1A). The Phy-
activated current was completely blocked by 50 nM
a-bungarotoxin (a-BGT; not shown), suggesting that
Phy binds to the agonist binding site of SAD/chick b2
nAChR. Neither Phy (Fig 1A) nor imidacloprid (not
shown) activated any current at the rat a7 nAChR.
The binding site for species-selective nAChR acti-
vation was determined by construction of chimeras of
rat a7 and Drosophila SAD cDNA, in which agonist
binding regions of a7 were replaced by the correspond-
ing sequence of SAD cDNA. The chimeras were
constructed using a two-step PCR ampli®cation. For
chimeras 7S.1 and 7S.2 the rat a7 cDNA region
His137 to Met226, including ACh binding loop B
with the putative Phy-site, and loop C were replaced,
whereas in 7S.2 the Phy-site (Lys165) was changed to
Gln. 7S.3 and 7S.4 contain the SAD sequence corre-
sponding to a7 His137 to Trp196 and a7 Trp196 to
Met226, respectively, including SAD loop C and loop
B. The ®nal PCR products were cloned into the
HindIII-XbaI sites of pcDNA3, were completely
sequenced, and expressed in Xenopus oocytes. Neither
chimeric nAChR. A. Phy (bar) induced a nicotinic inward current in oocytes
expressing SAD/chick b2 nAChR, in contrast to rat a7 nAChR. Holding
potential ÿ80mV. B. Oocytes were preincubated for 60 min in phosphate-
buffered saline (PBS), supplemented with 0.1% bovine serum albumin
(BSA), incubated for 120 min in PBS with 0.1% BSA containing 5 nM
[125I]a-BGT. After five washes radioactivity was counted. Non-specific
binding in non-transformed oocytes ranged from 9 to 22% of total binding,
and was subtracted to calculate specific [125I]a-BGT binding. For all
subtypes, specific binding was significantly higher compared to nonspecific
binding (P < 0.001).
ACh, Phy nor tetramethylammonium (TMA) was
able to induce an ion current in any of the chimeras.
To examine membrane expression of chimeric nAChR
proteins, binding of [125I]a-BGT was analysed in
single Xenopus oocytes. All chimeric cDNAs were
capable of assembly into homomeric nAChR, and
bound [125I]a-BGT (Fig 1B).
Subsequently, displacement of [125I]a-BGT by Phy
was analysed. Phy at 1 mM displaced 42 (7)% (n = 3)
of bound [125I]a-BGT from oocytes expressing 7S.3
(containing SAD loop C) and 55 (19)% (n = 4) from
wild-type SAD/chick b2 oocytes. Displacement was
less (13 (47)% (n = 6)) from oocytes expressing 7S.4
nAChR (containing SAD loop B with the putative
Phy-site).
CONCLUSIONS
The results show that Phy and imidacloprid selectively
activate insect nAChR. Further, the nAChR chimeras
of the rat a7 and Drosophila SAD a subunits success-
fully assemble into homomeric nAChR proteins, as
revealed by [125I]a-BGT binding. Lack of function of
the a7-SAD chimeras appears to be due to de®cient

1032 Pestic Sci 55:1027±1040 (1999)


coupling of agonist binding to gating of the ion
channel, indicating that agonist binding and subse-
quent ion channel opening are separate, but related
processes. Phy differentially displaces [125I]a-BGT
from the chimeric nAChR, suggesting that the b
subunit is not involved in Phy binding, and that Phy
targets the insect agonist binding loop C.
ACKNOWLEDGEMENTS
We would like to thank Dr B Schmitt (Max-Planck-
Institut fuÈr Hirnforschung, Frankfurt, Germany) and
Dr M Ballivet (University of Geneva, Switzerland) for
kindly providing us with cDNA encoding the SAD and
the chick b2 subunit, respectively.
This work was ®nancially supported by the US-EPA
(CR#82192701).
REFERENCES
1 Arias HR, Topology of ligand binding sites on the nicotinic
acetylcholine receptor. Brain Res Rev 25:133±191 (1997).
2 Zwart R, Oortgiesen M and Vijverberg HPM, Nitromethylene
heterocycles: selective agonists of nicotinic receptors in locust
neurons as compared to mouse N1E-115 and BC3H1 cells. Pest
Biochem Physiol 48:202±313 (1994).
3 Yamamoto I, Tomizawa M, Saito T, Miyamoto T, Walcott EC
and Sumikawa K, Structural factors contributing to insecticidal
and selective actions of neonicotinoids. Arch Insect Biochem
Physiol 37:24±32 (1998).
4 Van den Beukel I, Van Kleef RGDM, Zwart R and Oortgiesen M,
Physostigmine and acetylcholine differentially activate nicotinic
receptor subpopulations in Locusta migratoria neurons. Brain Res
789:263±273 (1998).
5 SchroÈder B, Reinhardt-Maelicke S, Schrattenholz A, McLane KE,
Kretschmer A, Conti-Tronconi BM and Maelicke A, Mono-
clonal antibodies FK1 and WF6 de®ne two neighboring ligand
binding sites on Torpedo acetylcholine receptor a-polypeptide. J
Biol Chem 269:10407±10416 (1994).
6 Palma E, Eusebi F and Miledi R, Co-expression of the neuronal a7
and L247T a7 mutant subunits yields hybrid nicotinic receptors
with properties of both wild-type a7 and a7 mutant homomeric
receptors. Proc Natl Acad Sci. USA. 94:1539±1543 (1997).
7 Bertrand D, Ballivet M, Gomez M, Bertrand S, Phannavong B
and Gundel®nger ED, Physiological properties of neuronal
nicotinic receptors reconstituted from the vertebrate b2 subunit
and Drosophila a subunits. Eur J Neurosci. 6:869±875 (1994).
Imidazenil, a new drug for the management of
convulsions in organophosphate intoxication
in rodents
Slawomir Rump,* Teresa Gidynska, Elzbieta Galecka,
Oktawiusz Antkowiak and Marek Kowalczyk
Department of Pharmacology and Toxicology, Military Institute of
Hygene and Epidemiology, 4 Kozielska Str, 01-163 Warsaw,
Poland
* Correspondence to: Slawomir Rump, Department of
Pharmacology and Toxicology, Military Institute of Hygene and
Epidemiology, 4 Kozielska Str, 01-163 Warsaw, Poland
Contract/grant sponsor: State Committee for Scientific Research;
contract/grant number: 14863/C–T00/95
(Received 25 August; revised version received 2 November 1998;
accepted 20 May 1999)
Pestic Sci 55:1027±1040 (1999)
Extended Summaries ± Neurotox '98
Abstract: The summary deals with the anti-con-
vulsant and antilethal effects of a new benzodiaze-
pine receptor partial agonist, imidazenil, in DFP
intoxication. It has been demonstrated that imida-
zenil (2±5 mg kgÿ1) signi®cantly decreases convul-
sion intensity, rapidly inhibits seizure patterns in
brain bioelectrical activity and signi®cantly in-
creases the anti-lethal ef®cacy of atropine plus
obidoxime therapy. These effects are comparable
to diazepam at 5 mg kgÿ1. However, diazepam ex-
hibits myorelaxant activity at therapeutic doses,
which are only observed at 5±10 times the therapeu-
tic doses of imidazenil.
Keywords: benzodiazepines; convulsions; DFP;
imidazenil; organophosphate
1 INTRODUCTION
Centrally mediated seizures are one of the toxic signs
following poisoning with organophosphates (OP),
Benzodiazepines (BDZ), especially diazepam, are very
effective in the management of these convulsions when
given as adjuncts to atropine and oxime.1,2 However,
sedation, myorelaxation and dependence make BDZ a
poor choice for the treatment in these states. Partial
BDZ receptor agonists are regarded as producing
these side effects only in very high doses. Some of these
drugs, like compound CGS9896, are very effective in
the treatment of OP intoxications3 but unfortunately
they are not registered as drugs. The recently reported
partial allosteric modulator of BDZ receptors, imida-
zenil (6-(2-bromophenyl)-8-¯uoro-4H-imidazo [1,5-
a] benzodiazepine-3-carboximide),4 is more like the
ideal drug for the management of OP-induced con-
vulsions. This drug is now under extensive studies in
many laboratories and clinics and is considered as a
potential novel anti-epileptic drug.5 The present study
was performed in order to determine anti-convulsant
effects of imidazenil in acute OP intoxications, as well
as to establish its antidotal ef®cacy in rodents.
2 MATERIAL AND METHODS
2.1 Tested drugs
Fluostigmine (diisopropyl phosphoro¯uoridate; DFP)
was used as a model OP compound
2.2 Anti-convulsant efficacy
2.2.1 Effects on convulsion intensity
These were determined on 20 Swiss strain male mice
divided in three groups (control and two experimental)
using a Convulsometer (Columbus Instruments,
USA). The control group received DFP (5 mg kgÿ1 sc)
and obidoxime (40 mg kgÿ1 ip) and experimental
groups additionally imidazenil (2 mg kgÿ1 ip) or diaze-
pam (5 mg kgÿ1) immediately after the intoxication.
Intensity of subsequent convulsions was measured 10,
30, 60 and 120 min after the treatment and presented
in g sÿ1
1033